Upload
others
View
8
Download
0
Embed Size (px)
Citation preview
LOW-POWER OP-AMP OPERATING IN
SUB-THRESHOLD REGION WITH
IMPROVED SLEW RATE
A Thesis submitted in partial fulfillment of the Requirements for the degree of
Master of Technology
In
Electronics and Communication Engineering
Specialization: VLSI Design & Embedded System
By
Apurbaranjan panda
Roll No. : 212EC2143
Department of Electronics and Communication Engineering
National Institute of Technology Rourkela
Rourkela, Odisha, 769 008, India
May 2014
Low-Power Op-Amp Operating In
Sub-threshold Region with
Improved Slew Rate
A Thesis submitted in partial fulfillment of the Requirements for the degree of
Master of Technology In
Electronics and Communication Engineering
Specialization: VLSI Design & Embedded System
By
APURBARANJAN PANDA
Roll No. : 212EC2143
Under the Guidance of
Prof. MUNSHI NURUL ISLAM
Department of Electronics and Communication Engineering
National Institute of Technology Rourkela
Rourkela, Odisha, 769 008, India
May 2014
Dedicated to…
My Dear Friends
DEPT. OF ELECTRONICS AND COMMUNICATION
ENGINEERING
NATIONAL INSTITUTE OF TECHNOLOGY, ROURKELA
ROURKELA – 769008, ODISHA, INDIA
CERTIFICATE
This is to certify that the work in the thesis entitled Low-Power Op-Amp Operating In
Sub-threshold Region with Improved Slew Rate by Apurbaranjan Panda is a record of
an original research work carried out by him during 2013 - 14 under my supervision and
guidance in partial fulfillment of the requirements for the award of the degree of Master of
Technology in VLSI Design & Embedded System. Neither this thesis nor any part of it,
to the best of my knowledge, has been submitted for any degree or diploma elsewhere.
Place:
Date:
Prof. M. N. Islam
Dept. of Electronics and Communication Engg.
National Institute of Technology
Rourkela-769008
DEPT. OF ELECTRONICS AND COMMUNICATION
ENGINEERING
NATIONAL INSTITUTE OF TECHNOLOGY, ROURKELA
ROURKELA – 769008, ODISHA, INDIA
DECLARATION I certify that
a) The work contained in the thesis is original and has been done by myself under the
general supervision of my supervisor.
b) The work has not been submitted to any other Institute for any degree or diploma.
c) I have followed the guidelines provided by the Institute in writing the thesis.
d) Whenever I have used materials (data, theoretical analysis, and text) from other
sources, I have given due credit to them by citing them in the text of the thesis and
giving their details in the references.
e) Whenever I have quoted written materials from other sources, I have put them
under quotation marks and given due credit to the sources by citing them and giving
required details in the references.
Apurbaranjan Panda
30th May 2014
i
ACKNOWLEDGEMENTS
It is my immense pleasure to avail this opportunity to express my gratitude, regards and
heartfelt respect to Prof. M. N. Islam, Department of Electronics and Communication
Engineering, NIT Rourkela for his endless and valuable guidance prior to, during and
beyond the tenure of the project work. His priceless advices have always lighted up my
path whenever I have struck a dead end in my work. It has been a rewarding experience
working under his supervision as he has always delivered the correct proportion of
appreciation and criticism to help me excel in my field of research.
I would like to express my gratitude and respect to Prof. K. K. Mahapatra, Prof. S. K.
Patra, Prof. S. Meher, Prof. D. P. Acharya, Prof. A. K. Swain and Prof. P. K. Tiwari for
their support, feedback and guidance throughout my M. Tech course duration. I would also
like to thank all the faculty and staff of ECE department, NIT Rourkela for their support
and help during the two years of my student life in the department.
I am also very thankful to all my classmates Prakash, Rakesh, Deepu, Naveen, Jayant,
Arvind, Kalpesh, Neel Kamal and seniors of VLSI lab especially Mr. Tom, Mr. Jaganath,
Mr. Debasish Nayak, Mr. Venkat Ramakrishna and all my friends who always encouraged
me in the Successful completion of my thesis work.
Apurbaranjan panda
ii
ABSTRACT
This thesis focuses on the weak inversion design of low power low voltage Op-amp.
The main objective of this thesis is to improve the slew rate of the Op-amp with a very less
amount of power consumption. For this purpose, an adaptive bias circuit is used to improve
the slew rate of an Op-amp in sub-threshold region by increasing the drain current of the
input stage transistor of Op-amp under the dynamic condition instead increasing the
quiescent current to reduce the power consumption. In this thesis we present sub-threshold
Op-amp circuit with two different adaptive bias topology. First, an adaptive biased circuit
with WTA (Winner Take All) topology, which is used in a class A Op-amp circuit and this
circuit designed using UMC 180nm technology with a supply voltage of 0.8V. The second
one is an adaptive bias circuit based on current subtractor topology, this topology is used in
a class AB Op-amp circuit with supply voltage of V. This thesis presents the
comparison between Op-amp with adaptive bias (both topology) and without adaptive bias
circuit.
iii
1.1 Motivation ...................................................................................................................................... 2
1.2 Literature Review .......................................................................................................................... 3
1.3 Overview of Thesis ........................................................................................................................ 4
2.1 Basics of Op-amp ........................................................................................................................... 6
2.2 Different Parameters Of Op-Amp ............................................................................................... 8
2.3 Ideal and Practical values of different Op Amp parameters ..................................................... 9
2.4 Different test circuits ................................................................................................................... 10
2.5 Conventional CMOS Op-amp .................................................................................................... 13
2.6 Simulation Results of Conventional Op-Amp ........................................................................... 15
3.1 Sub-threshold region operation of MOS transistor .................................................................. 20
3.1.1 Subthreshold current ................................................................................................................. 22
3.1.2 Transconductance of MOSFET in sub-threshold region .......................................................... 24
3.1.3 Capacitance of MOSFET in sub-threshold region .................................................................... 25
3.1.4 Leakage current in sub-threshold region .................................................................................. 26
CONTENTS
ACKNOWLEDGEMENTS ................................................................................... I
ABSTRACT ........................................................................................................ II
LIST OF FIGURES ............................................................................................ V
LIST OF TABLES ............................................................................................ VII
1 INTRODUCTION .......................................................................................... 1
2 BASICS OF OP-AMP .................................................................................. 5
3 LOW POWER OP-AMP ............................................................................. 19
iv
3.1.5 Small signal MOSFTE parameters in sub-threshold region ..................................................... 26
3.2 Op-amp operating in Sub-threshold Region ............................................................................. 27
3.3 Simulation Result ........................................................................................................................ 30
4.1 Adaptive Bias circuit ................................................................................................................... 35
4.1.1 WTA Adaptive Bias Circuit ..................................................................................................... 36
4.1.2 Op-amp with WTA circuit ........................................................................................................ 40
4.1.3 Simulation result for the Op-amp with WTA circuit ................................................................ 43
4.2 Adaptive bias circuit based on current subtractor topology ................................................... 45
1.1.1 Operation of current subtractor adaptive bias circuit ................................................................ 47
4.2.1 Simulation results of Op-amp with current subtractor adaptive bias topology ......................... 49
4.3 Simulation result comparison of different architecture of op-amp......................................... 52
REFERENCES ........................................................................................................................................ 56
APPENDIX A:.............................................................................................................................................. 58
APPENDIX B: ............................................................................................................................................... 60
4 SUB-THRESHOLD OP-AMP WITH IMPROVED SLEW RATE ................. 34
5 CONCLUSION ........................................................................................... 54
v
List of Figures
Figure 2-1 Basic Block Diagram of Op-amp ..................................................................................................... 7
Figure 2-2 Symbol of OP-AMP ....................................................................................................................... 11
Figure 2-3 Test Circuit for Open-Loop Gain .................................................................................................. 11
Figure 2-4 Test Circuit for CMRR .................................................................................................................. 12
Figure 2-5 Test circuit for PSRR ..................................................................................................................... 12
Figure 2-6 Test Circuit for Slew Rate .............................................................................................................. 13
Figure 2-7 Conventional CMOS Two Stage OP-AMP ..................................................................................... 14
Figure 2-8 Gain and Phase plot of conventional Op-amp ............................................................................... 16
Figure 2-9 CMRR Plot of Conventional Op-amp ............................................................................................. 16
Figure 2-10 PSRR Plot of Conventional Op-amp ............................................................................................ 17
Figure 2-11 Slew Rate of Conventional Op-amp ............................................................................................. 17
Figure 3-1 Magnitude of Inversion layer charge per unit area VS. Gate-Substrate Voltage ........................... 21
Figure 3-2 Logarithm of inversion layer charge per unit area VS. Gate substrate voltage ........................... 22
Figure 3-3 Output characteristics of an n-channel MOSFET operating in the sub-threshold region in a log-
linear plot ......................................................................................................................................................... 23
Figure 3-4 Drain Current Verses Gate Source Potential of A MOSFET in Saturation Region in Log Scale . 24
Figure 3-5 Two Stage Op-amp Operating in Sub-threshold Region ................................................................ 27
Figure 3-6 Gain and Phase Plot of Op-amp operating in subthreshold region ............................................... 31
Figure 3-7 CMRR of the Op-amp operating in the subthreshold region .......................................................... 31
Figure 3-8 Slew Rate of Op-amp operates in subthreshold region .................................................................. 32
Figure 4-1 Simple Block Diagram of an OP-AMP with Adaptive Bias Circuit ............................................... 35
Figure 4-2 Adaptive bias circuit using level shifter ......................................................................................... 36
Figure 4-3 Adaptive bias topology using WTA Block ..................................................................................... 37
Figure 4-4 Circuit diagram of a WTA topology ............................................................................................... 37
Figure 4-5 Schematic of WTA circuit ............................................................................................................... 40
Figure 4-6 Simulation result of WTA circuit .................................................................................................... 41
Figure 4-7 Schematic of Op-amp without current bias circuit ......................................................................... 42
Figure 4-8 Op-amp with WTA circuit .............................................................................................................. 42
vi
Figure 4-9 Gain and Phase plot of Op-amp with WTA circuit ......................................................................... 43
Figure 4-10 CMRR plot of Op-amp with WTA circuit .................................................................................... 44
Figure 4-11 Slew rate of Op-amp with WTA circuit ......................................................................................... 44
Figure 4-12 Current subtraction circuit .......................................................................................................... 46
Figure 4-13 Block diagram of Op-amp with subtractor circuit ....................................................................... 46
Figure 4-14 Two stage class AB Op-amp with subtractor circuit .................................................................... 47
Figure 4-15 Two stage OTA with adaptive bias circuit and gain enhancement circuit .................................. 49
Figure 4-16 Gain and Phase plot of Op-amp with subtractor adaptive bias circuit ........................................ 50
Figure 4-17 CMRR of Op-amp with subtractor adaptive bias circuit .............................................................. 50
Figure 4-18 Slew Rate of Op-amp with subtractor adaptive bias circuit ......................................................... 51
Figure 5-1 Layout of Conventional Op-amp .................................................................................................... 58
Figure 5-2 Layout of Sub-threshold region Op-amp ........................................................................................ 58
Figure 5-3 Layout of Op-amp with WTA circuit .............................................................................................. 59
Figure 5-4 Layout of Op-amp with subtractor adaptive bias circuit................................................................ 59
vii
List of Tables
Table 1: Comparison between Ideal and Practical parameters of Op-amp ..................................................... 10
Table 2: Simulation Results of Conventional Op-amp ..................................................................................... 18
Table 3: Small signal MOSFET parameters in sub-threshold region .............................................................. 26
Table 4: Simulation Results of Op-amp operating in sub-threshold region ..................................................... 33
Table 5: Simulation Results of sub-threshold Op-amp with WTA adaptive bias circuit ................................. 45
Table 6: Simulation Results of sub-threshold Op-amp with subtractor adaptive bias circuit ......................... 51
Table 7: Comparison of Op-amp performances ............................................................................................... 52
Table 8: W/L ratio of Conventional Op-amp .................................................................................................... 60
Table 9: W/L ratio of Op-amp operating in subthreshold region ..................................................................... 60
Table 10: W/L ratio of Op-amp with WTA circuit ............................................................................................ 61
Table 11: W/L ratio of Op-amp with subtractor based adaptive bias circuit ................................................... 61
1
1
INTRODUCTION
2
1.1 Motivation
The operational amplifier (Op-amp) is a really significant and widely used building
block in the analog and mixed signal circuit design field. The performance of Op-amp in
the integrated circuit is affects overall performance of the system. Proper design of Op-amp
is necessary according to the application requirement. The growing demand of portable
electronics devices makes the circuit designer think about low power low voltage
integrated circuit design. The tendency of bringing down the supply potential is a usual
tactic to lower the power consumption in digital circuit. Whereas, the supply reduction
subjects a number of challenges in the analog circuit design. As the technology shrinks the
transistor length scales down, power supply scales down, but the scaling rate of the
threshold voltage is not the same. Thus, the circuit behavior is obsoleted. Hence the design
of Op-amp has become an essential feature in order to provide efficient benefits by
technology scaling.
Working at very low supply voltage for an Op-amp a number of techniques were
developed such as MOSFTE Bulk-Driven technique, Floating Gate MOS, use of charge
pumps, and transistor operating in sub-threshold region. In sub-threshold region MOSFET
the current variation and transconductance is very similar as the Bipolar Junction
Transistor, in this region the small signal model is alike as in the saturation operation.
Current value reaches a lower level by sub-threshold design approaches, hence speed
limiting factor for an Op-amp could be observed in sub-threshold region. But for some
applications power consumption is vital factor rather than the speed. Applications like, in
biomedical signals where the amplitude of the signal is varies in some mV to uV range and
the frequency lies in some kHz range. Then sub-threshold operation can be utilize for such
application.
3
1.2 Literature Review
G. Palmisano et al, in 2001, [1] have defined a simple design process for two
stage CMOS operational amplifier for integrated into an analog design-based
CAD tool. They discussed about typical frequency compensation techniques for
high gain bandwidth product.
M. Alioto, in 2010, [2], presented a paper about the DC behavior of sub-
threshold CMOS logic for ultralow power circuit design perspective. They
provides an effective implementation of small signal as well as large signal
model for subthreshold region operation of MOS transistor.
A.J. Lopez-Martin,in 2005, [3], presented a paper describes about the methods
to reach low-voltage operational amplifiers with reduced power. In this paper he
presented three OTA structure with three different types of adaptive biasing
circuit. And all these three technique are implemented using the concept of level
shifter based adaptive biasing technique.
M. R. Valero et al. [4] presented a paper describes about low-power class-AB
Op-amp design. All transistors used in this design of Op-amp are biased in
subthreshold region. For improving the CMRR and slew rate of Op-amp
common mode feed-forward path and adaptive biased technique are employed
respectively.
4
H. Fathabadi, in 2013, [5] presented a paper on a low voltage and ultra-low
power operational transconductance amplifier (OTA). A current subtractor
circuit is used as adaptive biasing to improve the transient response of Op-amp.
David J. Comer et al.in 2004 [6], presented a paper based on the advantage of
weak inversion operation of transistor used in the differential stage to get higher
gain, less distraction and ease of compensation.
Luca Magneelli et al., in 2012, [7], have proposed the procedure of systematic
design of an Op-amp operating in the weak inversion region. And they found
that Op-amp in subthreshold region gives highest value of FOM for power and
load capacitor.
1.3 Overview of Thesis
This thesis presented a comparison study of low power Op-amp working in the sub-
threshold region with the conventional CMOS Op-amp working in saturation. This thesis
emphases the improvement of slew rate of sub-threshold Op-amp along with low power
consumption. The basics of Op-amp and a conventional Op-amp simulation result is
provided in Chapter 2 of the thesis. Chapter 3 describes about the sub-threshold behavior of
transistor and simulation of an Op-amp operating in the sub-threshold region is carried.
This chapter discussed about the problems appears in the sub-threshold Op-amp. Chapter 4
presents the solution to the problem occurred during the subthreshold operation of Op-amp.
In this chapter two new Op-amp structure with adaptive biased circuit are presented for
improving the slew rate of Op-amp. And also in this chapter we present comparison
between all the Op-amp circuits presented in this thesis. The conclusions are drawn in
chapter 5.
5
2
BASICS OF OP-AMP
6
Op-amp plays a vital role in analog and mixed signal circuit. It becomes one of
ingenious building blocks in analog circuit design. It was designed by John R. Ragazzini in
1947 to represent a unique type of amplifier that, by appropriate selection of its external
components. It could be organized in a collection of action, such as addition, subtraction,
integration, differentiation and amplification. The first application of Op-amp was in
analog computer [8]. Op-amps are the amplifiers that have sufficiently high forward open
loop gain, so that when op-amps are used in negative feedback, the overall close loop
transfer function becomes independent upon the forward path gain of the op-amp. This
principle has been used to develop many useful analog circuits. The primary requirement of
an op-amp is to have a very high open loop gain to implement the negative feedback
concept, very high resistance at the input and low impedance at the output stage. Now a
day the CMOS Op-amp circuits are frequently used in Integrated circuits. A two stage
unbuffered CMOS Op-amp is very popular among all design technique. This thesis focuses
mainly on various two stage unbuffered CMOS Op-amp design techniques.
2.1 Basics of Op-amp
Basically, in an Op-amp the input signal is a differential signal which is applied to a
differential amplifier as the input stage and the output stage is a single ended output, which
amplifies the difference of the two input given at the input stage of the Op-amp, hence
there is a differential to single ended conversion circuit must present in an Op-amp circuit.
As the output impedance of the Op-amp is very low and requires a high output voltage
swing a buffered stage is used at the output. To achieve close loop stability a compensation
circuit is required. The Figure 2-1 shows the basic block diagram of an Op-amp.
7
Figure 2-1 Basic Block Diagram of Op-amp
The differential amplifier is the input stage of the Op-amp. As its name suggests it
processes the difference between the two input signals. It terminates the common mode
voltage applied at the input terminals. As, it cancels most of the common voltage at the
input, hence, noise and bias voltages are negated out. Most of the electronic circuits use
differential amplifiers. But, practically some amount of common mode voltage factor is
always exists at the output of the differential amplifier [8].
The output of the differential amplifier is stated as:
(
) ((
) )
Ideally,
Where is the common mode gain of the amplifier and is the differential mode
gain of the amplifier.
By using a differential amplifier it is possible to cancel out noise and bias voltages that
seem on both inputs so a small common-mode gain is typically considered decent. The
common-mode rejection ratio is simply the ratio between differential-mode and common-
mode gain and it symbolizes the efficiency of the amplifier in rejecting voltages that are
8
common to both inputs from affecting the output. Common-mode rejection ratio (CMRR)
is given by:
For an ideal differential amplifier should be zero and CMRR should be infinite.
The differential amplifier used in CMOS technology gives a single ended output [9].
The out stage of two stage CMOS Op-amp is a common source which gives one more gain
stage in cascade with the differential gain stage. For design an Op-amp with very low
output impedance a source follower output stage is used which is treated as buffer
stage.The compensation circuit is improves the phase margin of the system and the system
becomes more stable. The desirable range of phage margin for a system is 45° to 60° [10].
2.2 Different Parameters Of Op-Amp
The performance of an Op-amp depends on the parameters associated with it, the
parameters are given as below:
DC Gain: it is the open loop gain of the Op-amp. Ideally DC gain of an Op-amp is
infinite, practically it is 60 dB to 100 dB
ICMR: It is defined as the range of common mode input voltage up to which the
transistors associated with the differential stage (first stage) are in saturation and gives a
constant gain.
CMRR: It is the ratio of differential mode gain to the common mode gain. An ideal Op-
amp should have very high CMRR value, practically CMRR is more than 60 dB. It is one
of the most important basic parameter of the Op-amp.
PSRR: The PSRR is defined as the product of ratio of the change in supply to change in
output voltage of the amplifier cause by the change in the power supply and open loop gain
of the Op-amp. It indicates how well the Op-amp doings to the deviation of power supply
9
(both positive and negative supply). Less is the variation of output of Op-amp
corresponding to the variation of power supply, higher is the PSRR. An Op-amp should
have very high PSRR. Practically this value is greater than 60 dB.
Slew Rate: It is defined as the maximum rate of change of output voltage w.r.to time. It
is expressed in terms of V/us. A good Op-amp should have high slew rate. It determines by
the charging and discharging of output capacitor by current available. Slew rate is
completely depends on the current sourcing or sinking capability of the input stage rather
than the output.
Settling time: It is the minimum time taken by the system to reach the steady state. For a
typical OP-amp this value is less than 1us.
Phase Margin: It is defined as the excess of phase required to make the system
marginally stable at the unit gain frequency. It is a measure of stability for close loop
system.
PM = 180+ ϕ
ϕ = Phase of system when gain is 0 dB.
Practically for a well-designed Op-amp this va e i d
For a becoming a system stable the phase margin and gain margin both are need to be
positive. However, the consideration of a phase margin is very vital because it affects the
transient response such as the settling time, slew rate, rise time, fall time and overshoot.
Input Offset Voltage: It defined as the voltage that must be applied between the two
input terminals of an Op-amp to null or zero the output voltage.
2.3 Ideal and Practical values of different Op Amp parameters
10
Here Table 1 shows the comparison of different Practical and Ideal parameters of an
OP-amp. Achieving ideal values of parameters for an Op Amp is very difficult with real
life constraints.
Table 1: Comparison between Ideal and Practical parameters of Op-amp
IDEALLY PRACTICALLY
RIN = infinite RIN is very high (in terms of MΩ)
ROUT=zero ROUT is a low value (in terms of KΩ)
DC GAIN = infinite DC GAIN ≥ dB
BANDWIDTH =infinite
BANDWITH is very high value (in terms
of MHz)
CMRR = infinite CMRR ≥ dB
PSRR = infinite PSRR ≥ dB
OFFSET = zero OFFSET ≠ ze (i de f mi iv t )
SLEW RATE = infinite SLEW RATE ≥ 5 V/
2.4 Different test circuits
For finding different characteristics parameter of an Op-amp various test circuit are
employed. This section presents schematic configuration of different test circuit are used in
this work.
11
Figure 2-2 Symbol of OP-AMP [8]
The Figure 2-2 shows a simple symbolic representation of a linear 3-terminal Op-amp.
Two input terminal are used for applying differential input signals into the Op-amp, the
negative signed input is known as inverting terminal of the input and positive one is the
non-inverting terminal. The Op-amp gives a single ended output and the output terminal is
shown in the symbol as out pin. The pin VDD and VSS are shown in the symbol generally
used as the terminals for the supply voltages. The op-amp is a dual power supplies
component in which the VDD terminal is used for positive power supply and VSS is used for
negative power supply, for some cases ground signal applied at the VSS terminal.
Figure 2-3 Test Circuit for Open-Loop Gain [10]
The Figure 2-3 shows the configuration for finding the open-loop gain of the Op-amp.
Here an AC signal VIN is applied to the non-inverting terminal and AC ground signal is
applied to the inverting terminal with supply voltage of VDD and VSS an AC analysis is
carried out. Then taking the ratio of output and input we can find the open-loop gain or DC
12
gain. By this configuration we can find out the UGB, phase margin and 3dB Band Width of
Op-amp too.
(1)
Figure 2-4 Test Circuit for CMRR [10]
The Figure 2-4 shows a simple configuration of finding the CMRR of an Op-amp. In
this case in a unit gain configuration of Op-amp two identical ac signal source Vcm are
connected in series with the inputs. An AC analysis is carried out for finding the CMRR of
the Op-amp. CMRR is found by the Equ.(2) [10].
(2)
Figure 2-5 Test circuit for PSRR [10]
The Figure 2-5 shows a configuration for finding out the PSRR of an Op-amp. In this
test circuit an ac signal vdd is connected in series with VDD to measure the PSRR. By an
AC analysis we can find out the PSRR of the Op-amp by the Equ. (3) [10].
13
(3)
Figure 2-6 Test Circuit for Slew Rate [10]
The configuration shown in the Figure 2-6 is a simple diagram of unit gain
configuration. This figure is used to find the slew rate and settling time of Op-amp by
applying a step signal at non inverting input of the Op-amp. Observing output waveform of
the Op-amp after a transient analysis we can find the SR and the settling time. From the
slope of the output signal during rise or fall time of output waveform we can determine the
SR.
2.5 Conventional CMOS Op-amp
A conventional two stage cmos Op-amp is shown in the Figure 2-7. The first stage of
this Op-amp is consist of M1-M5 and M8 transistors, the second stage is consist of M6-M7
transistors and miller compensation capacitor is used between the output and input of
the second stage. All transistors used in this conventional Op-amp are operates in
saturation region. The first stage is a differential stage, the differential signal is applied into
the pin and
the gate terminals of the transistor M1 and M2 respectively. These
transistor M1 and M2 are two NMOS driver transistors used to convert the differential
voltage to differential current, these differential currents are given to the current mirrored
14
load transistor M3 and M4. The transistor M5 and M8 are used to provide the biasing
current to the differential stage. The output of the differential stage is a single ended output
signal and it gives the difference of the input signals with a constant gain. The second stage
is a common source gain stage, the output of the first stage is applied to the gate of the M6
the transistor M7 is provides the bias current to the second stage and the output is taken
from the drain terminal of the M6 [10].
Figure 2-7 Conventional CMOS Two Stage OP-AMP
The DC gain of the first stage is
The DC gain of the second stage is
Overall gain of the Op-amp is [10]
15
( )( ) (4)
Slew rate of the conventional Op-amp is [9]
(5)
Where I5 is the current through the M5 transistor and it is the bias current of the input
stage.
The Gain bandwidth of the Op-amp is [9]
(6)
2.6 Simulation Results of Conventional Op-Amp
The conventional CMOS Op-amp shown in the Figure 2-7 is designed in UMC 180 nm
technology with supply voltage of 1.8 V and a bias current IDC of 20 uA. The simulation of
this Op-amp is done by CADENCE spectre circuit simulator.
The gain and the phase plot of the conventional Op-amp circuit is shown in the Figure
2-8. This graph is plotted by simulating the Op-amp configuration shown in Figure 2-7.
Here we applied a sinusoidal input signal of 10uV at a frequency of 1 KHz at the non-
inverting input of Op-amp. We found 64.37dB DC gain, 64.19° phase margin, 25.55 MHz
of UGB and a 15.94 KHz of 3dB band width for the conventional Op-amp. The power
consumption of the Op-amp found out from the transient analysis is 276.8 uW.
16
Figure 2-8 Gain and Phase plot of conventional Op-amp
Figure 2-9 CMRR Plot of Conventional Op-amp
17
The waveform of CMRR of the conventional Op-amp is shown in the Figure 2-9 and the
CMRR for this Op-amp 68.37dB. The PSRR of the Op-amp is 72.63dB shown in the
Figure 2-10.
Figure 2-10 PSRR Plot of Conventional Op-amp
Figure 2-11 Slew Rate of Conventional Op-amp
18
Slew rate and settling time of the Op-amp is found from the Figure 2-11 is 26.4 V/us
and 91.035ns respectively, for finding the SR a step signal increasing from 0V to 1.8V is
applied at the non-inverting terminal of a unit gain configuration of Op-amp.
The layout of the conventional Op-amp is shown in the Figure 5-1 and the post layout
simulation is done by CADENCE spectre layout simulator. The results of the post layout
simulation is shown in Table 2.
Table 2: Simulation Results of Conventional Op-amp
Parameters Schematic simulation
result
Post layout simulation
Result
DC Gain 64.37dB 64.39dB
CMRR 68.37dB 68.36dB
PSRR 72.63dB 72.65dB
UGB 25.55MHz 25.46MHz
SR 26.4V/us 26.3V/us
Phase margin 64.19° 66.63°
Settling time 91.035ns 98.614ns
Power 276.8uW 276.13uW
19
3
LOW POWER OP-AMP
20
With increasing the use of portable devices in medical science and wireless microsensor
nodes like many other application. It is becomes a challenge to increase the life time of
battery for those devices. Hence new circuit design technique needs to develop which
consumes less power for increasing the battery life time. Then low power and low voltage
circuit design is one alternate. As we know that Op-amp is a basic building block in the
circuit design so power consumption of Op-amp need to be reduced. Power consumption
can be decreases by reducing either total current in the circuit or supply voltage or by
reducing the both. As we decrease input current, though power dissipation is lowered, but
the dynamic range is affected and decreasing the supply voltage, it also becomes challenge
to keep transistors in saturation region. The major drawback on implementing low-voltage
CMOS circuits is the threshold voltage which does not scale down as the same rate as
compared to the power supply [11]. From the previous chapter the CMOS Op-amp
working in saturation region consumes power about 276.8uW. Transistor operating in sub
threshold region in an analog circuit consume less energy for active operation and leakage
power dissipation is less compare to the strong inversion region [7]. In the subthreshold
region, the drain current decreases but remains finite as drops several volt below .
The drain current reduces exponentially with . For low-voltage, low-power applications
Sub-threshold operation of the MOSFET can be used If an Op-amp works in sub-threshold
region the power consumption is less than working in saturation region.
3.1 Sub-threshold region operation of MOS transistor
Generally when a MOS transistor is operated in saturation region within any analog or
digital circuit, we assumed that the transistor is off (drain to source current is zero) when
the transistors gate to source potential is below threshold voltage of that transistor. But,
21
exactly below the threshold the drain current is exponentially proportional to the gate to
source potential [12] Thi c e t i k w “ bth e h d c e t”
The Figure 3-1 shows magnitude of inversion layer variation with gate to substrate
voltage. In this region the inversion charge is exponentially varies with gate potential, but
in strong inversion region the inversion charge linearly depends on gate potential this can
be observed clearly in the Figure 3-2. The depletion region charge is a small quantity,
because substrate of MOS transistor is a weakly doped region [13]. And in sub-threshold
region the inversion charge is very less than the source to drain region. Only diffusion
current flows in this region.
Figure 3-1 Magnitude of Inversion layer charge per unit area VS. Gate-Substrate
Voltage [14]
22
Figure 3-2 Logarithm of inversion layer charge per unit area VS. Gate substrate
voltage [14]
3.1.1 Subthreshold current
In subthreshold region the drain current [7]
(
) [ ( ) (
)] (7)
(
) (8)
Where
(9)
Where
(10)
23
Where n is the sub-threshold slope factor (
) [14], is the threshold
voltage of the transistor, is the gate voltage, is the source voltage, is the drain
voltage, is the thermal potential [
] , k is Boltzmann constant, T room
temperature in Kelvin [300K], q is charge of electron. is the oxide capacitance and is
mobility of electron.
If , then
<< 1, since e
-4 ≅ 0.018. The last term in Equ. (7) is
approaches equal to one, which can be ignored. The expression for drain current then
simplifies to:
( ) (11)
Figure 3-3 Output characteristics of an n-channel MOSFET operating in the sub-
threshold region in a log-linear plot [13]
The above Equ. (11) is independent on the value of , hence when in the
sub-threshold region, this region can be treated as saturation of MOSFET in sub-threshold.
24
As required to do so and it does not depend on as in the case above threshold. It is
easy to keep the MOSFET in saturation for subthreshold operation. This is very beneficial
for low-voltage designs.
Figure 3-3 shows that when ≅ mV the drain current saturated but in
triode region the drain current increases with increase in .
Figure 3-4 Drain Current Verses Gate Source Potential of A MOSFET in Saturation
Region in Log Scale [13]
The changes occurs in drain current as the gate to source voltage changes creates the
difference between above threshold and sub-threshold region operation of MOS transistor.
The current increases exponentially in a weakly-inverted MOSFET. Whereas the current
increases quadratircally (square law) in a strongly-inverted MOSFET. Figure 3-4 shows the
plot for with in logarithmic scale.
3.1.2 Transconductance of MOSFET in sub-threshold region
Transconductance of a subthreshold MOSFET is:
25
=
⁄ (12)
Equ. (12) shows a linear relationship between transconductance and drain current in
sub-threshold region. And the transconductance is independent upon device geometry
when the drain current constant. These two features of the sub-threshold region distinguish
from the strong inversion region, where the relation between transconductance and drain
current is a gives a square-law and also for the device geometry.
The subthreshold MOSFETs behaves very much like BJT transistor, both drain current
and transconductance is very much similar as of BJT. A bipolar transistor has a
transconductance of =
, which is equivalent to the expression for a subthreshold
MOSFET if we set n=1. For a MOS transistor current through its gate is zero, but in a
bipolar transistor always there is a small amount of current pulls by its base. This can make
circuit design much simpler.
3.1.3 Capacitance of MOSFET in sub-threshold region
Intrinsic capacitors present in a MOSFET are due to the charge stored in the channel.
Each of these capacitors are dependent upon the fraction of the total gate oxide capacitor
according to the bias voltage. The extrinsic capacitors are the source and drain
junctions capacitors and the gate overlap capacitors to source and drain diffusions [13].
Their values are basically independent of the drain current.
In weak inversion, most of the intrinsic capacitor values are very small if the channel is
small. The gate-to-substrate capacitance made by the series connection of and the
depletion capacitance is the only non-negligible capacitance .it is given by [13]:
(13)
And is smaller than the (total gate oxide capacitance).
26
3.1.4 Leakage current in sub-threshold region
In saturation region the only dominating leakage current is due to the DIBL (Drain
Induced Barrier Lowering) factor [13]. The effect of DIBL on can be taken in to account
on the expression as follows [2].
(14)
The values of and ,
Where is the body to source voltage, is the DIBL effect compnent and is
substrate effect component. The DIBL effect is associated to the distance between drain
and source region i.e. the is mainly depends on L.
3.1.5 Small signal MOSFTE parameters in sub-threshold region
From the above discussion and equations the small signal parameters (transconductance
, drain to source resistance , the substrate trance conductance and DC gain ) of a
transistor in the sub-threshold region are reported in the Table 3.
Table 3: Small signal MOSFET parameters in sub-threshold region
PARAMETERS IN SUB-THRESHOLD REGION
[
]
27
3.2 Op-amp operating in Sub-threshold Region
In the chapter 2 we have seen that the conventional Op-amp is consume power in a
range of hundreds of uW. The power of the Op-amp is depends upon the bias current of the
input as well as the output stage. The bias current is the drain current of the transistor used
for biasing. We have seen that the drain current of the transistor is touches a lower value
when operates in sub-threshold region, the small signal model of Op-amp in sub-threshold
is very similar that of the saturation region and
value reaches the maximum value in this
region [13] . Hence an Op-amp designed in sub-threshold region will consume less power
as compared to the saturation region operation.
Figure 3-5 Two Stage Op-amp Operating in Sub-threshold Region
The Op-amp shown in the Figure 3-5 is a two stage Op-amp configuration operating in
the sub-threshold region where PMOS transistors are used as the input of the differential
28
stage. Because the noise is five time less in PMOS transistor than NMOS transistor
[10]. All the transistors are used in this Op-amp are operates in the weak inversion region.
The transistor M1- M5 and M8 are the first stage and M6-M7 are used as the second stage
of the Op-amp. Where M1 and M2 are the driver transistor which converts the differential
input to differential current, M3 and M4 are forms current mirrored load to convert the
differential current to single ended output voltage and the transistor M5 gives the bias
current to input stage and the capacitor Rc and the transistor makes a frequency
compensation network for achieving close-loop stability of the Op-amp. The transistor
is used instead of the resistor required for the RC compensation network.
The DC gain of the Op-amp shown in the Figure 3-5 is very similar to the DC gain is
conventional Op-amp
( ) ( )
The value of the is the output resistance of the transistor
Then
( )( )
(15)
The (15) shows that the gain of the Op-amp in sub-threshold region is depends on the
values of and the sub-threshold slope factor. The sub-threshold slope factor for the
transistor is a constant (1 < < 2). As depends upon the channel length, L is the only
design parameter to set the gain of the Op-amp in the sub-threshold region. The bias
29
current of the differential pair can be set by the required input referred noise spectral
density [7]. The of the input stage transistor also fixed with fixing the bias current.
The gain-bandwidth product (GBW) of the Op-amp is
And
(16)
The slew rate of the Op-amp is
(17)
From the (15) it is seen that the DC gain of the sub-threshold Op-amp is independent
upon the drain current it is only depends upon value of L whereas the SR and GBW of the
Op-amp is dependents upon the value of the drain current. Hence the SR of the Op-amp is
reduced while operating in sub-threshold region.
The design of Op-amp in sub threshold region for specific parameters is carried out as
follows [7]:
From the specified value of DC gain we can set the value of L for transistor used in the
Op-amp.
For achieving essential input referred noise power spectral density the bias current of
the differential stage can set
( )
Where is the white noise component which is generally available at the high
frequency, for low frequency a gate referred flicker noise component is added to the inputs.
30
From (12) the transconductance of the first stage is automatically fix with fixing the
value of . From the drain current equation the aspect ratio of the transistor M1,2 is set
for a suitable value, which makes the transistor to work in weak inversion region.
(
)
(
)
The compensation can be set either from the value of GBW or from the value of SR.
As the is fixed from the desire value of GBW we can find the value of as follows
The bias current of the second stage is fixed for getting the maximum output swing in
the output of the Op-amp. After the fixing the value of the bias current in the second stage
we can set the value of aspect ratio of by using the drain current equation.
For setting the (compensation resistance) for the RC compensation network we have
to consider the load capacitance, parasitic capacitance at the output of first stage and the
desired phase margin as for the design strategy given in [1]
( √
( )
)
As in this Op-amp we replaced the resistor with transistor , we can set the aspect
ratio of the transistor and its gate potential for matching the of the transistor with the
value of found from the above equation.
3.3 Simulation Result
The simulation of the Op-amp schematic shown in the Figure 3-5 is carried out in
CADENCE circuit simulator with a supply voltage of 0.8V and bias current of 140nA
applied to the node. The Op-amp is designed in UMC 180nm technology and channel
length of 720nm.
31
Figure 3-6 Gain and Phase Plot of Op-amp operating in subthreshold region
Figure 3-7 CMRR of the Op-amp operating in the subthreshold region
The Figure 3-6 is shows the gain and phase plot of the Op-amp described in the previous
section, from this plot we have seen that the gain of the Op-amp is 67.79dB, UGB is
1.224MHz,3dB band width is 328.5KHz and phase margin is 63.6 . The power
consumption of this circuit is found by the transient analysis is 830.48nW.
32
The Figure 3-7 is shows the CMRR plot of the Op-amp shown in the Figure 3-5, the
CMRR of this Op-amp is 56.91dB. The PSRR of this Op-amp is found by using the test
circuit shown in the Figure 2-5 is 34.51dB.
For finding the slew rate and the settling time of Op-amp we have applied a pulse signal
of two voltage level 0 and 400mV, rise time and fall time of 1ns and the period the signal is
100us with pulse width of 50us to a unit gain configuration and the output of this
configuration is shown in the Figure 3-8. The slew rate and the selling time of the Op-amp
are 0.3116V/us and 5.28 us respectively.
Figure 3-8 Slew Rate of Op-amp operates in subthreshold region
The layout of the Op-amp operating in sub-threshold region is shown in the Figure 5-2
and the post layout simulation is done by CADENCE spectre layout simulator. The results
of the post layout simulation is shown in Table 4.
33
Table 4: Simulation Results of Op-amp operating in sub-threshold region
Parameters Schematic simulation
result
Post layout simulation
Result
DC Gain 67.79dB 67.7dB
CMRR 56.91dB 57.62dB
PSRR 34.51dB 34.47dB
UGB 1.224MHz 1.183MHz
SR 0.311V/us 0.308V/us
Phase margin 63.60° 60.6°
Settling time 5.28us 5.41us
Power 830.48nW 830.51nW
34
4
SUB-THRESHOLD OP-AMP WITH
IMPROVED SLEW RATE
35
A power efficient and high slew rate conventional Op-amp design in the sub - threshold
region is quite difficult because the charging and discharging of the load capacitor depends
upon the biased current of the input stage. And the slew rate of an Op-amp is always
depends upon the bias current too. If the bias current somehow increased so that of total
power of the amplifier is not permitted to increase then the slew rate limitation can be
improved. An adaptive bias circuit used to remove the bias current limitation for a
differential signal applied at the input stage. A block diagram of an Op-amp with adaptive
bias circuit is presented in the Figure 4-1.As the adaptive bias circuit is used to increase the
slew rate of Op-amp by increasing the biasing current under the dynamic condition without
changing the quiescent current of the Op-amp gives the benefit of low power consumption.
Figure 4-1 Simple Block Diagram of an OP-AMP with Adaptive Bias Circuit
4.1 Adaptive Bias circuit
The adaptive bias circuit is provided the bias current depending upon the difference of
the input signal applied at the differential stage of the Op-amp. Various types of adaptive
bias circuits are discussed in [15, 16, 4, 5] .
36
4.1.1 WTA Adaptive Bias Circuit
The WTA (winner takes all) circuit design is based on the adaptive biasing topology
using two level shifter [3]. The adaptive biasing circuit using two level shifter is shown in
the Figure 4-2. Here two driver stage transistor of an Op-amp is cross coupled by two DC
level shifter.
Figure 4-2 Adaptive bias circuit using level shifter [3]
When the gate to source voltage of both the transistor are equal i.e.
(Quiescent condition). Then both the transistor M1 and M2 carry the same amount of DC
current by . If decreases the source potential of the M1 transistor is decreasing, but
the voltage at the source of the M2 is remains unchanged. So the current through M1
decrease and the current through M2 increases. The current reaches the maximum value
which is very high than the DC current under the quiescent condition. To drive the low
impedance source terminals of M1 and M2 transistor very low impedance level shifters are
required. And the level shifter circuit should have the capability to produce a large current
during the charging and the discharging period of load capacitor.
37
Figure 4-3 Adaptive bias topology using WTA Block [3]
The Figure 4-3 shows a diagram of adaptive bias topology using WTA Block, WTA
b ck tp t i w y the m xim m “wi e ” f the i p t v t ge The ef ected t the
common source node of M1 and M2 is a maximum of input voltages shifted by the
constant DC voltage . When and
are equal i.e. under quiescent condition the
common source terminal gives maximum value relates the common mode voltage of
the input signals. Therefore , and the quiescent currents through the
driver transistors are depends and controlled by in the similar way it works in the circuit
shown in Figure 4-2.
Figure 4-4 Circuit diagram of a WTA topology [3]
38
Under the dynamic condition when a difference occurs between the input signals, if the
decreases, that means it is lower than
then the common source terminal tracks
higher input voltage i.e. higher than the common mode voltage of the inputs. .
Therefore, the resulting source to gate voltage is larger than previous topology for the
same input voltage, and a larger dynamic current boost up is attained.
The Figure 4-4 shows a circuit diagram of a WTA topology. Here for the transistor M1,
M1A, M2 and M2B all the gate and source terminals are tied together. Under the quiescent
condition when a common mode voltage is applied (
) the voltage at A_vdd
will be , the current follows is the quiescent current .
When decreases increases then the drain current of M1A transistor increases, so
the drain voltage of M1A is also increasing. As the drain of the M1A is connected to the
gate of M2A transistor the current through M2A decreases that reduces the potential at the
node A_vdd, but on the other side the increase that increases the node potential of
A_vdd, Therefore the potential of A_vdd is set by , the aspect ratio of M1B and the
current , i.e. . Similarly, when
decreases below the node voltage
of A_vdd becomes .
[
] (18)
For V,
, and currents and are
[
] (19)
*
+ (20)
39
[
] (21)
(
) (22)
Then
* (
)
+ (23)
[
] (24)
And
(25)
In a similar way, for V, , and currents and are
*
+ (26)
[
] (27)
And
(28)
Hence from the equation (7), (8), (10) and (11) we have seen that by using the WTA
circuit the current through M1 and M2 increases exponentially with under the
dynamic condition.
Hence
( *
+ ) and ( *
+ )
40
For a fully balance input ac signal the node A_vdd is an ac ground and
. The differential current for small signal condition is similar to
the simple Op-amp structure because, due to use of WTA circuit the largest of the input
transistors is .
4.1.2 Op-amp with WTA circuit
We discussed about WTA topology in the previous section which can used as an
adaptive bias circuit in an Op-amp circuit for improving the SR of the Op-amp by
increasing the tail current under the dynamic condition.
Figure 4-5 Schematic of WTA circuit
The Figure 4-5 shows the schematic of the WTA circuit the transistor M12 and M13 are
used as the current source, which provides the bias current under the quiescent condition
41
for that these two transistor are operates in the saturation region. As we discussed the WTA
circuit always tracks to the higher value of the input signal we can observe that from the
plot shown in Figure 4-6.
Figure 4-6 Simulation result of WTA circuit
Figure 4-6 shows the result of WTA circuit, when any one of the input signal or
goes high the voltage at node a_vdd grasps maximum value and it gives minutest value
when both inputs are the same.
The Figure 4-7 shows an Op-amp circuit without any bias current circuit, the
configuration of this Op-amp is very similar to the Op-amp discussed in the section 3.2 but
in this case the node a_vdd of this Op-amp is connected to a_vdd node of the WTA circuit,
here the a_vdd node of WTA circuit is connected to the common source terminal of the
driver transistor at the differential stage. The source and gate of the transistor M8 and M9
are connected with transistor M1 and M2 of the input stage of the Op-amp respectively.
42
Figure 4-7 Schematic of Op-amp without current bias circuit
Figure 4-8 Op-amp with WTA circuit
The complete structure of Op-amp with WTA structure is shown in Figure 4-8. Here all
the transistors are in sub-threshold region except M12 and M13. In this circuit three bias
voltage supply (cbias, nbias and bias) required for operating the circuit properly. The
“ bi ” bi ed v t ge de i c ected t the M12 d M13 th t m ke the e tw
transistor to work in saturation region as well as for providing the bias current to the
43
ci c it Th gh the de “cbi ” bi v t ge i pp ied t the t i t that makes
the transistor work as a compensating resistor, which is connected in series with the
compensating capacito f m ki g RC c mpe ti etw k The de “bi ” i ed
to provide a biased current to the second stage of the Op-amp.
4.1.3 Simulation result for the Op-amp with WTA circuit
The design of the circuit shown in the Figure 4-8 is done by using the UMC 180nm
technology with a supply voltage of 0.8V and the simulation of the circuit is carried out by
using Cadence circuit simulator ADE (analog design environment).
Figure 4-9 Gain and Phase plot of Op-amp with WTA circuit
44
Figure 4-10 CMRR plot of Op-amp with WTA circuit
Figure 4-11 Slew rate of Op-amp with WTA circuit
The Figure 4-9 is shows the gain and phase plot of the Op-amp operating in sub-
threshold region with adaptive bias circuit, from this plot we have seen that the gain of the
Op-amp is 67.67dB, UGB is 1.325MHz, 3dB band width is 284.7 KHz and phase margin is
45
58.1 . The power consumption of this circuit is found by the transient analysis is
1.002uW.The Figure 4-10 is shows the CMRR plot of the Op-amp shown in the Figure 4-8,
the CMRR of this Op-amp is 81.46dB. The PSRR of this Op-amp is found by using the test
circuit shown in the Figure 2-5 is 34dB. For finding the slew rate and the settling time of
Op-amp we have applied a pulse signal of two voltage level 0 and 400mVto a unit gain
configuration and the output of this configuration is shown in the Figure 4-11. The slew
rate and the selling time of the Op-amp are 1.69V/us and 4.6us.
The layout of the Op-amp operating in sub-threshold region is shown in the Figure 5-3
and the post layout simulation is done by CADENCE spectre layout simulator. The results
of the post layout simulation is shown in Table 5
Table 5: Simulation Results of sub-threshold Op-amp with WTA adaptive bias circuit
Parameters Schematic simulation
result
Post layout simulation
Result
DC Gain 67.67dB 67.66dB
CMRR 81.45dB 81.463dB
PSRR 34.0dB 33.9dB
UGB 1.325MHz 1.269MHz
SR 1.636V/us 1.695V/us
Phase margin 58.1° 55.0°
Settling time 4.66us 4.63us
Power 1.0027uW 1.0042uW
4.2 Adaptive bias circuit based on current subtractor topology
An adaptive biased circuit based on subtractor is circuit is presented in this section
[5]. The Figure 4-12 shows the circuit diagram of the subtractor circuit.
46
Figure 4-12 Current subtraction circuit [5]
The current and are related to the signal applied to the differential stage of the Op-
amp. Here current flows through M1 and is mirrored by the M1 and M2 current mirror
pair circuit. Current flows through M3 is and the mirrored transistor M4 gives the
current with a gain of A and the current at the M4 transistor becomes ( ).
Figure 4-13 Block diagram of Op-amp with subtractor circuit
The Figure 4-13 shows a block diagram of an Op-amp with subtractor circuit as adaptive
biased circuit. Two subtractor blocks are used to provide improvement in the bias current
under the dynamic condition. This subtractor adaptive biasing circuit is generally used in a
class AB output stage type Op-amp circuit.
47
Figure 4-14 Two stage class AB Op-amp with subtractor circuit
The circuit diagram shown in Figure 4-14 is a simple class AB stage OTA (operational
transconductance amplifier) with two subtractor circuit. The shaded region shows two
subtractor circuit connected to solve the current limitation for improving the slew rate. The
transistor M1-M10 are form a conventional OTA circuit, all other transistors are used for
making current subtractor circuit.
1.1.1 Operation of current subtractor adaptive bias circuit
In the Figure 4-14 the current flows through M1 and M3 and this current is mirrored in
M11, M12 and M16. The current flows through M2 and M4 and it is mirrored in M13,
M14, and M21. current flows through M10 and mirrored in M9. The drain of M9 is
connected to the common source terminal of the M1 and M2. Under the quiescent
condition when differential signal is zero i.e. the tail current becomes .
When differential voltage i.e. then increases more than . Hence
maximum current flows through M1. The transistor M1, M3, M12, M20 and M19 creates a
positive feedback loop that further increases the current through M1 and reduces the
current through M2 and it becomes off. Therefore, becomes zero and increases to
. The current is mirrored in M11, M16 and the current through M17 becomes
48
negative hence the transistor M17 and M18 becomes off. And current flows through
M20 and the current through M19 will be . As the tail current becomes
Further this current returned and increased with gain of A through M1,M3 in the feedback
loop created by M1, M3, M29, M20 and M19. Repetitively this course continues, hence
the tail current becomes.
( ) ∑
Where A is the ratio of aspect ratio of M18,19 to M17,20
When A becomes greater than one the tends to infinity, therefore A must be less than
1. Then
When A=0 this circuit is not behave like adaptive bias i.e. total tail current becomes .
Hence we have to keep the aspect ratio of M18,19 always less than the aspect ratio of
M17,20.
The first stage gain of OTA shown in the Figure 4-13 is
a very small value which
affect the overall gain of the amplifier. To increase the gain another gain enhanced network
is added to the circuit as shown in the Figure 4-15.
49
Figure 4-15 Two stage OTA with adaptive bias circuit and gain enhancement circuit [5]
The load transistor of the in Figure 4-14 is a diode connected load, hence the output
resistance of load is very small. The shaded region shown in the Figure 4-15 is a
compensator circuit used to increase the DC gain. By adding the extra circuit the trance
conductance of the overall load is becoming , which increases the gain of the
first stage as well as the overall gain.
4.2.1 Simulation results of Op-amp with current subtractor adaptive bias
topology
The simulation of sub-threshold Op-amp with current subtractor adaptive bias topology
has been done using UMC 180 nm technology with supply voltage of V. The current
source shown in the Figure 4-15 is replaced by a transistor with a bias voltage.
The Figure 4-16 shows gain, UGB and phase margin are 51.1dB, 4.322MHz and 57.70
respectively.
50
Figure 4-16 Gain and Phase plot of Op-amp with subtractor adaptive bias circuit
Figure 4-17 CMRR of Op-amp with subtractor adaptive bias circuit
The CMRR of the Op-amp is 88.49dB as shown in the Figure 4-17. The PSRR of this
Op-amp is 80.23dB. This Op-amp consumes 1.39uW of power. The SR and the settling
51
time of the Op-amp is found out by applying a step signal the unit gain configuration and
the output is shown in the Figure 4-18. The SR is 4V/us and the settling time is 0.871us.
Figure 4-18 Slew Rate of Op-amp with subtractor adaptive bias circuit
The layout of the circuit is shown in Figure 5-4 and the post layout simulation result is
tabulated in Table 6.
Table 6: Simulation Results of sub-threshold Op-amp with subtractor adaptive bias
circuit
Parameters Schematic simulation
result
Post layout simulation
Result
DC Gain 51.1dB 51.1dB
CMRR 88.49dB 88.47dB
PSRR 80.23dB 80.23dB
UGB 4.322MHz 4.524MHz
SR 4.065V/us 3.41V/us
Phase margin 57.70° 63.20°
Settling time 0.871us 0.909us
Power 1.39uW 1.42uW
52
4.3 Simulation result comparison of different architecture of op-amp
In this thesis, we discussed about four different type Op-amp circuit, in this section we will
show the comparison of these four Op-amp performances.
All four Op-amp circuit simulation has been done in UMC 180nm technology with BISM
3v3 level 11 model, the simulation results of schematic and av-extracted view has been
presented in previous sections. The performance comparison of all four Op-amp structure
is presented in the Table 7.
Table 7: Comparison of Op-amp performances
PARAMETERS Conventional Sub-threshold
With WTA
circuit based
adaptive biased
circuit
With subtractor
based adaptive
bias circuit
SUPPLY
VOLTAGE 1.8V 0.8V 0.8V 0.6V
DC gain 64.37dB 67.79 dB 67.67 dB 51.1 dB
UGB 25.55MHz 1.224 MHz 1.325 MHz 4.322 MHz
SR 26.4 0.311 1.63 4.065
CMRR 68.37dB 56.91 dB 81.45 dB 88.49 dB
PSRR 72.63 dB 34.51 dB 34.0 dB 80.23 dB
SETTLING
TIME 91.035 5.28 4.66 0.909
POWER 276.8 W 0.830 W 1.0027 W 1.39 W
PHASE
MARGIN 64.19° 63.60° 58.1° 57.7°
OPERATION
MODE Saturation Sub-threshold Sub-threshold Sub-threshold
The basic objective of this work was to reduce the power of an Op-amp with a desirable
SR value. From the above table we can observe that a conventional Op-amp working in
saturation region gives a good frequency response, desirable DC gain, CMRR, PSRR,
system stability and very high slew rate. But this Op-amp is very power hungry circuit,
53
which contradicts our objective of Op-amp design. For reducing the power of the Op-amp
we preferred to go for an Op-amp operating in sub-threshold region. While, working with
sub-threshold region it reduces the power of the circuit as per our requirement, but, it
reduces the slew rate and increases the settling time with a descent phase margin. The
reduction of slew rate is needs to be improved, for that we used the adaptive bias circuit.
We examined two adaptive bias techniques(WTA, subtractor) and the results are presented.
The Op-amp using WTA circuit is gives five times higher slew rate than the sub-threshold
Op-amp circuit. the power of this Op-amp is little higher than the sub-threshold Op-amp.
The gain and phase margin values are almost similar for both Op-amp. The Op-amp circuit
with subtractor adaptive biased topology gives a better slew rate value than two above
discussed sub-threshold Op-amp circuit. But gain of this Op-amp circuit is lower than the
all other Op-amp presented in this thesis.
54
5
CONCLUSION
55
An Op-amp operating in sub-threshold region gives the advantage low power over the
conventional Op-amp in saturation region, the power required for the conventional Op-amp
presented in this thesis is 276 W, but in sub-threshold Op-amp consumes only 0.830 W
of power. As the drain current in sub-threshold region reduced the slew rate of
subthreshold Op-amp is reduced from 26.4 to 0.311 . For improving the slew
rate we have used adaptive biasing technique which increases the biased current of the
subthreshold Op-amp under the dynamic condition by keeping the biased current lower for
quiescent condition. Two different topology of adaptive biased technique are used. Firstly,
adaptive biased using WTA circuit which provides a maximum gate to source voltage
difference when difference between the input signal increases. By using this method we
achieved a slew rate of 1.63 with a power consumption of 1.0027 W. This Op-amp
gives a decent the DC gain and phase margin. Secondly, adaptive bias circuit using current
subtractor circuit, in this technique the bias current is increased by a factor of ( )
(where, A<1) under the dynamic condition. By using this adaptive biased technique the
slew rate is increased to 4.065 and the power consumption is becomes 1.39 W and it
gives a better frequency response over WTA adaptive bias technique, but DC gain of this
Op-amp is reduced.
56
REFERENCES
[1] G. Palmisano, G. Palumbo and S. Pennisi, "Design procedure for two-stage CMOS
transconductance operational amplifiers: a tutorial," Analog Integrated Circuits and
Signal Processing, no. 27, pp. 179-189, 2001.
[2] M. Alioto, "Understanding DC behavior of subthreshold CMOS logic through closed-
form analysis," IEEE Transactions on Circuits and Systems I, vol. 57, no. 7, pp. 1597-
1607, 2010.
[3] A. J. López-Martín, S. Baswa, J. Ramírez-Angulo and R. González- Carvajal, "Low-
voltage super class AB CMOS OTA cells with very high slew rate and power
efficiency," IEEE J. Solid-State Circuits, vol. 40, no. 5, pp. 1068-1077, 2005.
[4] M. R. Valero, S. Celma, B. Calvo and C. Azcona, "An ultra low-power low-voltage
class AB CMOS fully differential OpAmp for Long-Life Autonomous Portable
Equipment," IEEE Transactions On Circuits And Systems—II, vol. 59, no. 10, 2012.
[5] H. Fathabadi, "Ultra Low Power Improved Differential Amplifier," Springer science
Circuits Syst Signal Process, vol. 32, pp. 861-875, 2013.
[6] D. J. Comer and D. T. Comer, "Using the Weak Inversion Region to Optimize Input
stage Design of CMOS Op amp," IEEE Transaction on Circuit and System-II, vol. 51,
no. 1, pp. 8-14, 2004.
[7] L. Magnelli, F. A. Amoroso, F. Crupi, G. Cappuccino and G. Iannaccone, "Design of a
75-nW, 0.5-V subthreshold complementary metal–oxide–semiconductor operational
amplifier," INTERNATIONAL JOURNAL OF CIRCUIT THEORY AND
APPLICATIONS Wiley Online Library, 2012.
[8] S. Franco, Design with Operational Amplifire and Analog integrated Circuit, New
57
York: Tata McGraw Hill, 2002.
[9] B. Razavi, Design of a Analog CMOS Integrated Circuit, New Delhi: Tata McGraw-
Hill, 2012.
[10] P. E. ALLEN and D. R. HOLBERG, CMOS ANALOG CIRCUIT DESIGN, I. S.
Edition, Ed., Oxford: Oxford University press, 2009.
[11] M. Razzaghpour and A. Golmakani, "An Ultra-low-voltage Ultra-Low-Power OTA
With Improved Gain-Bandwidth Product," in International Conference on
Microelectronics, 2008.
[12] R. R. Troutman, "Subthreshold Slope for Insulated Gate Field-Effect transistors,"
IEEE Trans. Electron Devices, vol. 22, pp. 1049-1051, 1975.
[13] A. Wang, B. Highsmith and A. P. Chandrakasan, Sub-Threshold Design for Ultra
Low-Power Systems, New York: Springer-Verlag, 2010.
[14] Y. Tsividis, Operation_and_Modeling_of_the_MOS transistor, New York: OXFORD
UNIVERSITY.
[15] H. Y. Huang, W. R. B and L. J. C, "High-gain and high-bandwidth rail-to-rail
operational amplifier with slew rate boost circuit," in IEEE Int. Symp. Cir. and Sys.,
2006.
[16] A. R. Kim, H. R. Kim, . Y. Park, . Y. K. Choi and . B. S. Kong, "Low-power class-AB
CMOS OTA with high slew rate," in Int. SOC Design Conf. (ISOCC), 2009.
58
APPENDIX A:
LAYOUTS
Figure 5-1 Layout of Conventional Op-amp
Figure 5-2 Layout of Sub-threshold region Op-amp
59
Figure 5-3 Layout of Op-amp with WTA circuit
Figure 5-4 Layout of Op-amp with subtractor adaptive bias circuit
60
APPENDIX B:
W/L RATIO
Table 8: W/L ratio of Conventional Op-amp
Transistor W/L
M1,M2 6
M3,M4 14
M5,M8 12
M6 173
M7 75
Table 9: W/L ratio of Op-amp operating in subthreshold region
Transistor W/L
M1,M2 5/0.72
M3,M4 3/0.72
M5,M8 1.2/0.72
M6 1/0.72
M7 1
MC 1.2/2
61
Table 10: W/L ratio of Op-amp with WTA circuit
Transistor W/L
M1,M2 5/0.72
M3,M4 3/0.72
M6 1/0.72
M7, M8, M9 1
MC 1.2/2
M10, M11 40/0.72
M12,M13 10/0.72
Table 11: W/L ratio of Op-amp with subtractor based adaptive bias circuit
Transistor W/L
M1,M2 1/0.54
M3,M4 0.8/0.54
M5, M8 21.5/0.54
M6, M7 1
M9, M10, M11, M12, 0.72/0.54
M15, M16, M17, M20 0.72/0.54
M18, M19 .7/0.54
M21, M22 0.72/0.54
M23, M24 1